If you wanted to generate random numbers from whatever the null distribution is, you would first generate a uniform random number and then apply the inverse cdf of the null distribution to get a random value. That value would be a (one-sided) p-value.alan2 said:Help. I need an intuitive, non-mathematical explanation of why p-values from hypothesis testing are uniformly distributed. I was talking to a social scientist and got a blank stare. I couldn't come up with anything except the proof. Thanks.
alan2 said:Help. I need an intuitive, non-mathematical explanation of why p-values from hypothesis testing are uniformly distributed.
tnich said:If you wanted to generate random numbers from whatever the null distribution is, you would first generate a uniform random number and then apply the inverse cdf of the null distribution to get a random value. That value would be a (one-sided) p-value.
A P-value is a statistical measure that helps to determine the likelihood of obtaining a particular result or more extreme results, given that the null hypothesis is true. It is used in hypothesis testing to determine the significance of the results.
When P-values are uniformly distributed, it means that they are evenly spread out across all possible values between 0 and 1. This indicates that the null hypothesis is true and there is no significant difference between the observed data and the expected results.
The uniform distribution of P-values is used to assess the significance of the results in a statistical analysis. A P-value that falls within the expected range of a uniformly distributed set of values supports the null hypothesis, while a P-value that is outside of this range suggests that the null hypothesis should be rejected.
No, P-values cannot be used to determine causation. They only provide evidence for or against the null hypothesis, but they do not prove causation. To establish causation, other factors such as controlled experiments and alternative explanations must be considered.
No, uniformly distributed P-values are not always reliable. They can be affected by factors such as sample size, study design, and the presence of confounding variables. Therefore, it is important to interpret P-values in the context of the specific study and to consider other factors before drawing conclusions.